Peptides. Vol. 11, pp. 707-712. ©Pergamon Press plc, 1990. Printed in the U.S.A.
0196-9781/90 $3.00 + .00
A Monoclonal Antibody to Somatostatin With Potent In Vivo Immunoneutralizing Activity H E L E N C. W O N G , J O H N H. W A L S H , t H O N G Y A N G , YVETTE TACHt~ AND ALISON M. BUCHAN*
Center for Ulcer Research and Education, West Los Angeles Veterans Administration Center Brain Research Institute, and Department of Medicine, University of California, Los Angeles, CA 90073 *Department of Physiology, University of British Columbia, Vancouver, Canada R e c e i v e d 30 M a r c h 1990
WONG, H. C., J. H. WALSH, H. YANG, Y. TACHI~ AND A. M. BUCHAN. A monoclonalantibody to somatostatin withpotent in vivo immunoneutralizingactivity. PEPTIDES 11(4) 707-712, 1990.--The spleen from a Robertsonian mouse with high titer and affinity antiserum after being immunized with somatostatin-14 conjugated to keyhole limpet hemocyanin was fused with FOX-NY cells. Hybridomas were cloned by limiting dilution, subcloned, and ascites was produced from the highest affinity clone in pristine-primed Balb/c mice. Ascites fluid contained approximately 20 mg/ml IgG and bound 50% of 1 fmol ~25I-[Tyr~]-somatostatin at a final dilution of 1:10,000,000. Binding of this IgG antibody, CURE. $6, was inhibited by 50% at 40 pM concentrations of either somatostatin-14 or somatostatin-28, but was not inhibited by [D-Trp8]-somatostatin at 1000-fold higher concentrations. The antibody produced very intense specific immunohistochemical staining of somatostatin endocrine cells in the stomach and pancreas and of intestinal somatostatin neurons with extremely low background staining. Intravenous injection of 2 mg purified antibody in urethane-anesthetized rats resulted in 300-fold increase in plasma GH within 15 min. CURE. $6 is a high affinity monoclonal antibody directed at the biologically active somatostatin ring structure. This antibody is useful for in vivo immunoneutralization of exogenous and endogenous somatostatin in the rat and also is an excellent reagent for immunohistochemical localization of somatostatin. Somatostatin Monoclonal antibody Growth hormone release Paracrine actions Immunoneutralization
SOMATOSTATIN inhibits release of growth hormone from the pituitary and also inhibits release of many gastrointestinal hormones (17,21). It has multiple inhibitory actions on gastrointestinal secretory processes, often by stimulating the cAMP inhibitory GTP binding protein, Ni (13). Delivery of somatostatin from cells that produce the peptide to target cells can occur by endocrine, paracrine, or neurocrine routes (10). Elucidation of physiologically important endogenous actions of somatostatin requires a specific blocking or neutralizing agent. Suitable receptor antagonists have not yet been developed. Immunoneutralization with polyclonal antibodies has been utilized successfully in vivo to study the role of somatostatin in regulation of GH release (11) and of gastric acid secretion (16). Monoclonal antibody immunoneutralization of endogenous somatostatin also has been used to study regulation of gastric acid secretion in rats (14,15). Passive immunization with polyclonal antisera may produce varying responses depending upon the biological properties of the antisera. Relatively high doses of antisera usually are needed for in vivo immunoneutralization. Development of high affinity mono-
Gastric secretion
Gastrin
Immunohistochemistry
clonal antibodies as immunoneutralizing reagents can provide virtually unlimited amounts for biological studies. These antibodies can be extensively characterized as to biological properties in vivo as well as in vitro. In the present report we describe the production and characterization of this antibody and its use for in vivo immunoneutralization studies and for immunostaining. The high affinity and activity of this somatostatin monoclonal antibody under in vivo conditions make it a useful reagent for exploring the physiologic actions of this peptide. METHOD
Peptides Synthetic [Tyr~]-somatostatin, [Tyr°]-somatostatin, [Tyr~]somatostatin, somatostatin-14, somatostatin-28, [D-Trp8]-soma tostatin and [D-Trpll]-somatostatin, CGRP, and CGRP(23-37) were obtained from Peninsula Laboratories. Cyclic somatostatinrelated peptide, octreide (Sandostatin®), was a gift from Sandoz.
IRequests for reprints should be addressed to John H. Walsh, M.D., Department of Medicine, UCLA Medical School, 10833 Le Conte Avenue, Los Angeles, CA 90024.
707
708
Another somatostatin analog, cyclic Cys-Phe-Phe-dTrp-Lys-ThrPhe-Cys (ODT8-SS), was a gift from D. R. Rivier, Salk Institute. Synthetic gastrin was from Research Plus and pentagastrin from Ayerst.
Animals and Cell Lines Robertsonian mice containing the translocated 8.12 chromosome (18) were obtained from Jackson Laboratory, Bar Harbor, ME. Male Sprague-Dawley rats and Balb/c mice were obtained from Charles River, North Wilmington, MA. FOX-NY cell lines were obtained from Hyclone Laboratory, Logan, UT.
Reagents Basic culture medium consisted of 15% fetal bovine serum (FBS, obtained from Hyclone Sterile Systems, Inc.) in RPMI1640 containing 25 mM HEPES buffer (M. A. Bioproducts), and 1% L-glutamine-penicillin-streptomycin solution (Irvine Scientific) at pH 7.4. Special media included adenine and thymidine added to basic culture medium. AT medium contained 1.5 × 10- 4 M adenine hydrochloride (Sigma) and 3.2 x 10 5 M thymidine (Sigma). AAT medium was prepared from AT medium by adding aminopterin (Sigma) to a final concentration of 10 7 M. Red blood cell lysis buffer contained 0.1 g KHCO 3, 0.83 g NH4C1, and 4 mg EDTA dissolved in 100 ml deionized water. Polyethylene glycol (J. T. Baker Chemical Co.) was dissolved in serum-free basic culture medium to a final concentration of 60% PEG in the presence of 5% CO2. Cell freezing medium consisted of 10% dimethyl sulfoxide (DMSO, Mallinckrodt) in fetal bovine serum. All solutions were prepared under sterile conditions and stored at 4°C.
Immunization of Mice Somatostatin-14 (5 rag) was conjugated to 2.5 mg keyhole limpet hemocyanin (Calbiocbem) by addition of 20 mg Nethyl-N(3 dimethyl-amino-propyl) carbodiimide (ICN Biochemicals) in 0.05 M phosphate-buffered saline, pH 7.4, and reacted for 5 hr at 22°C. Six 10-week-old Robertsonian mice were immunized at intervals of 6 to 8 weeks by 0.5 ml intraperitoneal injections of somatostatin conjugate emulsified with equal parts of complete Freund's adjuvant, equivalent to 50 Ixg somatostatin per dose. Three weeks after each immunization mice were bled from tail veins, and binding to 125I-[Tyr~]-somatostatin was determined by radioimmunoassay. All mice developed antibody after the second immunization, with titers ranging from 1:10,000 to 1:100,000. Sensitivity was determined by radioimmunoassay, using somatostatin-14 as standard. The mouse with highest sensitivity combined with high titer was used for fusion.
WONG, WALSH, YANG, TACHI~ AND BUCHAN
cells were resuspended in AT medium at a density of 2.5 × 106 cells/ml and plated on microtiter plates at 2 drops from a 5-ml pipette per well. After overnight incubation at 37°C in 5% CO 2, 2 drops AAT medium were added to each well. Culture supernatant was aspirated and cells were fed with AAT medium on alternate days for 10 days, then cells were fed with AT medium at least twice before being cultured in basic culture medium. Each well was screened by liquid phase radioimmunoassay for antibody production, and positive wells were expanded for cloning. Aliquots of master cultures were frozen in PEG/FBS and stored in liquid nitrogen. Cloning was done by limiting dilution with the help of thymocyte feeder layers. Thymocytes were obtained from 6week-old Balb/c mice under sterile conditions and suspended in basic culture medium. Cells from a single thymus gland were suspended in 20 ml medium, and 2 drops of cell suspension were added to each well of 96-well microtiter culture plates. Hybridoma cells from the master culture were diluted sufficiently to produce average cell numbers of 10, 5, 2, and 0-1 per 2 drops of medium, and each of these concentrations was plated onto separate microtiter plates that contained thymus feeder layers. Wells were examined under the microscope after 3-5 days incubation to determine number of viable cells, and supernatants were removed and tested for antibody activity by radioimmunoassay after l-hr incubation at 37°C of 1:5 and 1:50 dilutions in pH 7.4 PBS containing 10% proteinate and 1500 cpm ~25I-[TyrJ]-somatostatin plus 2500 units Trasylol. Wells that contained only one viable hybridoma cell and positive antibody activity were expanded separately. After expansion, supernatants were examined for antibody activity. Sensitivity was estimated by comparing binding in the presence and absence of 100 pM noniodinated somatostatin. Supernatants with binding activity that was inhibited in the presence of cold hormone were recloned and expanded separately. Clones that produced highest sensitivity antibody were grown into mass culture for ascites production. Ascites was produced in 10-week-old BALB/c mice that had been primed one week earlier by intraperitoneal injection of 0.5 ml pristine (2,6,10,14-tetramethyl pentadecane, Sigma) by intraperitoneal injection of 1-2 x 107 hybridoma cells in 0.5 ml serum-free culture medium. Mice were examined daily from 7 to 14 days after injection for production of ascites, which was removed by paracentesis. Ascites fluid was used for further characterization of monoclonal antibodies.
Antibody Characterization Antibodies were characterized for sensitivity and specificity by radioimmunoassay as previously described (23). Immunoglobulin class and subclass were characterized in ascites fluid by radial immunodiffusion, using antibodies to mouse immunoglobulins provided by ICN Immunobiologicals.
Fusion and Antibody Production
Hormone Assays
Fusion of spleen cells and FOX-NY cells, at a ratio of 5:1, was achieved with PEG. The spleen was removed aseptically and infused with 10 ml 15% FBS-RPMI-1640 medium to liberate lymphocytes. Cells were centrifuged at 500 x g twice for 10 min, washed in serum-free medium, exposed to RBC lysis buffer, resuspended in medium containing 15% FBS, centrifuged and resuspended in 50 ml 15% FBS-RPMI for fusion. FOX-NY cells were grown to concentrations of 4-6 x 105 cells/ml prior to fusion. Spleen cells and myeloma cells were mixed, centrifuged, resuspended in 1 ml 60% PEG over 1 min followed by serial additions of medium over 5 min, with stirring. After centrifugation, fused
GH assay was performed at the Salk Institute as previously described (20). Gastrin radioimmunoassay was performed using antibody 1611, specific for the C-terminal biologically active portion of gastrin as previously described (7).
Purification of Monoclonal Antibody Somatostatin antibody was purified from 2 ml ascites fluid through a protein-A Sepharose CL-4B column (Pharmacia) equilibrated with 1.5 M glycine/3 M NaC1, pH 8.9, and eluted with 0.1 M citrate buffer, pH 3.0. Fractions containing IgG were pooled
MONOCLONAL ANTIBODY TO SOMATOSTATIN
709
BIF -
B/F 4.0-
13
5.512
5.0Sandoslatin
2.510
~ ¢ ~
.~
-- --I .
.
.
.
~
2.0-
~ssoOT88918803
'
~ c G R P 1-37 09-
08-
07-
06
Plole I
PIo~e E
Plote ]]I 05
FIG. 1. Binding of labeled somatostatin by supernatants from individual
wells of master culture 96-well plates at 5th day after initial fusion of spleen cells. The well from Plate II with highest binding was subcloned. Supernatant diluted 1:50.
04
03
and dialyzed against 4 liter 0.15 M NaC1. Protein content of fractions was measured by Bio-Rad protein microassay.
GH Release in Anesthetized rats Male Sprague-Dawley rats weighing 200-250 g were housed under controlled temperature and 12-hr light/dark cycle conditions. Animals were allowed to consume chow and water ad lib, then deprived of food but not water during the 24 hr prior to experiments. Animals were anesthetized with urethane, 1.5 g/kg IP. For studies of GH release, a jugular catheter was utilized to inject antibody and to obtain serum samples. Two baseline samples were obtained at 15-min intervals, then 2 mg purified monoclonal antibody or control monoclonal antibody against glycine-extended gastrin (1) was injected intravenously. Four additional blood samples were obtained at 15-min intervals, and the volume removed was replaced by saline. Serum from these samples was assayed for GH as previously described (20).
Immunohistochemistry Immunohistochemical studies were performed as previously described (2), using monoclonal antibody CURE.S6 at dilutions of 1:100-1:10,000 as the first reagent and FITC-conjugated antimouse IgG as the second reagent. Specificity was determined by inclusion of 10 nM somatostatin in the first incubation. Tissues were fixed in Bouin's solution containing 1% glacial acetic acid for 2 hr, followed by 2 changes of 70% ethanol before paraffin embedding. Sections were cut at 6 micron thickness for histological analysis.
Statistical Analysis Differences between means were analyzed by use of t-tests for unpaired values, and p values